A5067892749- Machine Learning in Materials Science
- X-ray Diffraction in Crystallography
- Computational Drug Discovery Methods
- Physics of Superconductivity and Magnetism
- Theoretical and Computational Physics
- Advanced Battery Materials and Technologies
- Organic and Molecular Conductors Research
- Scientific Computing and Data Management
- Various Chemistry Research Topics
- Advanced X-ray and CT Imaging
- Membrane Separation and Gas Transport
- X-ray Spectroscopy and Fluorescence Analysis
- Electron and X-Ray Spectroscopy Techniques
- Magnetic properties of thin films
- Computational Physics and Python Applications
- Advancements in Battery Materials
- Radiomics and Machine Learning in Medical Imaging
- Perovskite Materials and Applications
- Analytical Chemistry and Chromatography
- Electronic and Structural Properties of Oxides
- Spectroscopy and Quantum Chemical Studies
- Advanced battery technologies research
- SARS-CoV-2 detection and testing
- Age of Information Optimization
- Advanced Multi-Objective Optimization Algorithms
Brookhaven National Laboratory
2019-2025
Columbia University
2019-2021
University of Rochester
2014
Vidant Medical Center
2003
East Carolina University
2003
University of Miami
1998
Abstract X-ray absorption spectroscopy (XAS) produces a wealth of information about the local structure materials, but interpretation spectra often relies on easily accessible trends and prior assumptions structure. Recently, researchers have demonstrated that machine learning models can automate this process to predict coordinating environments absorbing atoms from their XAS spectra. However, are difficult interpret, making it challenging determine when they valid whether consistent with...
Simulations of excited state properties, such as spectral functions, are often computationally expensive and therefore not suitable for high-throughput modeling. As a proof principle, we demonstrate that graph-based neural networks can be used to predict the x-ray absorption near-edge structure spectra molecules quantitative accuracy. Specifically, predicted reproduce nearly all prominent peaks, with 90% peak locations within 1 eV ground truth. Besides its own utility in analysis inference,...
Abstract The unparalleled success of artificial intelligence (AI) in the technology sector has catalyzed an enormous amount research scientific community. It proven to be a powerful tool, but as with any rapidly developing field, deluge information can overwhelming, confusing, and sometimes misleading. This make it easy become lost same hype cycles that have historically ended periods scarce funding depleted expectations known AI winters. Furthermore, although importance innovative,...
As machine learning (ML) methods continue to be applied a broad scope of problems in the physical sciences, uncertainty quantification is becoming correspondingly more important for their robust application. Uncertainty-aware have been used select applications, but largely scalar properties. In this work, we showcase an exemplary study which neural network ensembles are predict x-ray absorption spectra small molecules, as well pointwise uncertainty, from local atomic environments. The...
X-ray absorption near-edge structure (XANES) spectroscopy is a robust and element-specific tool for probing the atomic of materials. Traditional methods work in forward direction by simulating XANES spectra from models. Here, authors present opposite: computational method predicting local structural geometry which so-called inverse problem solved using supervised machine learning. The robustness fidelity are demonstrated an average 86% classification accuracy K-edge hundreds materials across...
Abstract X-ray absorption spectroscopy (XAS) is a premier technique for materials characterization, providing key information about the local chemical environment of absorber atom. In this work, we develop database sulfur K-edge XAS spectra crystalline and amorphous lithium thiophosphate based on atomic structures reported in Chem. Mater ., 34, 6702 (2022). The simulations using excited electron core-hole pseudopotential approach implemented Vienna Ab initio Simulation Package. Our contains...
Improved understanding of structural and chemical properties through local experimental probes, such as X-ray absorption near-edge structure (XANES) spectroscopy, is crucial for the design functional materials. In recent years, significant advancements have been made in development data science approaches automated interpretation XANES structure–spectrum relationships. However, existing studies primarily focused on crystalline solids small molecules, while fewer efforts devoted to disordered...
Accurate classification of molecular chemical motifs from experimental measurement is an important problem in physics, chemistry, and biology. In this work, we present neural network ensemble classifiers for predicting the presence (or lack thereof) 41 different on small molecules simulated C, N, O K-edge X-ray absorption near-edge structure (XANES) spectra. Our not only achieve class-balanced accuracies more than 0.95 but also accurately quantify uncertainty. We show that including multiple...
Electrochemical degradation of solid electrolytes is a major roadblock in the development solid-state batteries. Combining X-ray absorption spectroscopy characterization, first-principles simulations, and machine learning, here we report atomic-scale oxidative mechanisms sulfide using Li3PS4 (LPS) as model system. The begins with decrease Li neighbor affinity to S atoms, followed by formation S-S bonds PS4 tetrahedron deforms. After first cycle, motifs become strongly distorted, PS3 start...
Abstract Spectroscopy techniques such as x-ray absorption near edge structure (XANES) provide valuable insights into the atomic structures of materials, yet inverse prediction precise from spectroscopic data remains a formidable challenge. In this study, we introduce framework that combines generative artificial intelligence models with XANES spectroscopy to predict three-dimensional disordered systems, using amorphous carbon ( -C) model system. work, new based on diffusion model, recent...
<a:math xmlns:a="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><a:msub><a:mrow><a:mi>Zn</a:mi><a:mi>Cl</a:mi></a:mrow><a:mn>2</a:mn></a:msub></a:math> solutions are promising electrolytes for aqueous zinc-ion batteries. Here, we report a joint computational and experimental study of the structural dynamic properties <d:math xmlns:d="http://www.w3.org/1998/Math/MathML"...
We study polarons in the one-dimensional Bond-Peierls electron-phonon model which phonons on bonds of a lattice modulate hopping electrons between sites and contrast results to those known for breathing-mode Peierls problem. By inspecting atomic limit, we show that polaronic dressing mass enhancement depend momentum dependence phonons. For dispersionless phonons, are perfectly localized unlike their counterparts because carrier creates string phonon excitations can only be annihilated via...
A semisupervised machine learning method for the discovery of structure-spectrum relationships is developed and then demonstrated using specific example interpreting x-ray absorption near-edge structure (XANES) spectra. This constructs a one-to-one mapping between individual descriptors spectral trends. Specifically, an adversarial autoencoder augmented with rank constraint (RankAAE). The RankAAE methodology produces continuous interpretable latent space, where each dimension can track...
Carbone et al., (2023). Lightshow: a Python package for generating computational x-ray absorption spectroscopy input files. Journal of Open Source Software, 8(87), 5182, https://doi.org/10.21105/joss.05182
The Anderson Impurity Model (AIM) is a canonical model of quantum many-body physics. Here we investigate whether machine learning models, both neural networks (NN) and kernel ridge regression (KRR), can accurately predict the AIM spectral function in all its regimes, from empty orbital, to mixed valence, Kondo. To tackle this question, construct two large databases containing approximately 410k 600k functions single-channel impurity problem. We show that NN models with point-wise mean...
A fully microscopic model of the doping-dependent exciton and trion linewidths in absorption spectra monolayer transition metal dichalcogenides low temperature low-doping regime is explored. The approach based on perturbation theory avoids use phenomenological parameters. In regime, we find that linewidth relatively insensitive to doping levels, while increases monotonically with doping. On other hand, argue shows a somewhat stronger dependence. magnitudes are likely be masked by phonon...
We employ variational autoencoders to extract physical insight from a dataset of one-particle Anderson impurity model spectral functions. Autoencoders are trained find low-dimensional, latent space representation that faithfully characterizes each element the training set, as measured by reconstruction error. Variational autoencoders, probabilistic generalization standard further condition learned promote highly interpretable features. In our study, we variables strongly correlate with well...
Effective response to natural or man-made disasters (i.e., terrorism) is predicated on the ability communicate among many organizations involved. Disaster exercises enable disaster planners and responders test procedures technologies incorporate lessons learned from past exercises. On May 31 June 1, 2002, one such exercise event took place at Camp Lejeune Marine Corps Base in Jacksonville, North Carolina. During exercise, East Carolina University tested: (1) in-place Telehealth networks (2)...
We generalize the family of approximate momentum average methods to formulate a numerically exact, convergent hierarchy equations whose solution provides an efficient algorithm compute Green's function particle dressed by bosons suitable in entire parameter regime. use this approach extract ground-state properties and spectral functions. Our approximation-free framework, dubbed generalized cluster expansion (GGCE), allows access exact numerical results extreme adiabatic limit, where many...